JP4513819B2 - Video conversion device, video display device, and video conversion method - Google Patents

Description

  The present invention relates to a video processing apparatus, and more particularly to a technique for creating an interpolation frame from a frame in a video signal and performing frame rate conversion.

  In recent years, attention has been focused on the application of frame rate conversion as a technique for improving image quality in order to improve unnatural movements such as blurring and rattling in moving image display by increasing the number of frames of an image. However, in order to obtain a clear improvement effect for improving the moving image performance as described above, it is necessary to generate an interpolation frame with high accuracy. As an advanced interpolation method that has been frequently used, there is a method that uses a motion compensation process using a motion vector between frames using the current frame and the previous frame. In this method, a motion between frames is specified based on motion vector information, and an interpolation frame is created. In order to obtain a highly accurate interpolation frame, it is necessary to increase the accuracy of the motion vector. For example, a search method divided into two stages (for example, Non-Patent Document 1) or a comparison with surrounding motion vectors A technique such as performing a smoothing process for removing an inappropriate motion vector by performing this technique has been proposed (see, for example, Patent Document 1).

Japanese Patent Laid-Open No. 2002-27414 (Section 9, FIG. 9) Toshiba Review Vol.59 No.12 (2004)

  However, in the conventional motion compensation processing, it is difficult to obtain an accurate motion vector in an image with multiple motions in one screen or in a motion where multiple moving objects intersect. It will be bankrupt.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a technique capable of detecting a motion vector more accurately and performing frame rate conversion with high image quality.

To achieve the above object, the present invention may be configured as described in the appended claims.

  According to such a configuration, in the frame rate conversion process using the motion compensation method, the adaptive interpolation control that captures the characteristics of the video realizes the improvement of the moving image performance and the video with the occurrence of the failure suppressed. It becomes possible.

  According to the present invention, it is possible to detect a motion vector more accurately, and thus it is possible to perform frame rate conversion with high image quality.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing an example of the configuration of an image processing apparatus according to the first embodiment of the present invention.

  In FIG. 1, 1001 is an input signal, 1002 is a resolution conversion unit, 1003 is a frame rate conversion (hereinafter referred to as FRC: Frame Rate Conversion) unit, 1004 is an image memory, 1005 is a timing control unit, 1006 is a display unit, 1007 Is an FRC conversion mode signal.

  The resolution conversion unit 1002 performs an enlargement / reduction process on the input image to an image format adapted to the display unit 1006. The FRC unit 1003 performs frame rate conversion according to the FRC conversion mode signal 1007 set from the outside, and the timing controller unit 1005 displays the frame rate conversion output obtained from the FRC unit 1003 on the display unit 1006. Adjust the timing.

  FIG. 2 shows an example of the configuration of the FRC unit 1003 in FIG.

  In FIG. 2, 2001 is the current frame signal, 2002 is the signal one frame before, 2003 is the FRC conversion mode signal, 2004 is the motion vector detection unit, 2005 is the motion vector, 2006 is the image feature detection unit, 2007 is the determination signal, 2008 is An interpolation frame generation unit, 2009 is an interpolation frame signal, 2010 is a frame rate conversion output signal, and 2011 is a memory interface unit.

  The motion vector detection unit 2004 detects a motion vector 2005 from the current frame signal 2001 and the previous frame signal 2002. As a detection method, any of the block matching method, gradient method, phase correlation method and the like described in the prior art may be used. Here, N × N (N is an integer) as shown in FIG. The case where the block matching method is used will be described below. In FIG. 3, 3001 is the current frame signal, 3002 is the previous frame signal, 3003 is the interpolated frame signal, 3004 is the target block, 3005 is the motion vector search range, 3006 is the minimum difference value block pair, and 3007 is the motion vector.

  On the interpolation frame 3003, a search range of a predetermined number of blocks is provided with respect to the target block 3004 to be interpolated, centering on a block at a point-symmetrical position in each of the current frame signal 3001 and the previous frame signal 3002. In the case of FIG. 3, 11 horizontal blocks and 5 vertical blocks are set. The block pair 3006 with the smallest difference value is detected by the difference value matching between the current frame signal 3001 at the position of the point symmetric pair in the time direction and the block on the previous frame signal 3002 around the target block 3004, The direction is output as a motion vector 3007.

  The obtained motion vector 2005 (3007) is input to the interpolation frame generation unit 2008. The interpolation frame generation unit 2008 calculates the pixel value of the target block in the interpolation frame as the average value of the pixel values of the minimum difference value pair.

  In FIG. 3, a single interpolated frame is inserted at the center of gravity position temporally with respect to the frame at the original input frame rate (hereinafter referred to as a key frame) as in the case of conversion from 60 Hz to 120 Hz. Explained when to do. On the other hand, FIG. 4 shows an example in which a plurality of interpolation frames are inserted between key frames, such as conversion from 24 Hz to 60 Hz of a movie. It is the FRC conversion mode signal 1007 (2003) that determines the operation mode of frame rate conversion, such as whether to convert from 60 Hz to 120 Hz and from 24 Hz to 60 Hz.

  In FIG. 4, the same components as those of FIG. 3 are denoted by the same reference numerals, and the description thereof is omitted.

  4001 is motion vector search range 1, 4002 is motion vector search range 2, 4003 is interpolation frame 1, 4004 is interpolation frame 2, 4005 is target block 1, 4006 is target block 2, 4007 is motion vector 1, 4008 is motion vector 2.

  In FIG. 4, a motion vector search range 1 4001 is set on the current frame signal 3001 and a motion vector search range 2 4002 is set on the previous frame signal 3002 for the target block 1 4005 on the interpolation frame 1 4003. As can be seen from the figure, the size of the search range is motion vector search range 1 <motion vector search range 2. This is because the temporal distance from the current frame signal 3001 of the interpolation frame 1 4003 is shorter than the temporal distance from the signal 3002 one frame before, and as a result, the difference value between the blocks at the point-symmetrical positions. The size of the search range that can be set during the matching operation is different. Similarly, for the target block 2 4006 on the interpolated frame 2 4004, the motion vector search range 2 4002 is displayed on the current frame signal 3001 and the motion vector is displayed on the previous frame signal 3002 due to the difference in the temporal center of gravity. Search range 1 4001 must be set. Similar to the conversion from 60 Hz to 120 Hz, the difference value matching calculation is performed within the search range, the block pair having the smallest difference value is calculated, and the direction thereof is output as the motion vector 2007. Here, since there are two interpolation frames, two motion vectors (4007 and 4008) are necessary. The interpolation frame generation unit 2005 calculates the interpolation pixel value in consideration of the temporal centroid position of each of the interpolation frame 1 4003 and the interpolation frame 2 4004. Specifically, in FIG. 4, the pixel values of the target block 1 4005 on the interpolation frame 1 and the target block 2 on the interpolation frame 2 4006 (referred to as I (1) and I (2), respectively) in the time direction The following equations (1) and (2) are respectively calculated by load addition considering the position of the center of gravity.

(Equation 1) I (1) = (3 * Y1_a + 2 * Y0_a) / 5
(Equation 2) I (2) = (Y1_b + 4 * Y0_b) / 5
FIG. 5 is a diagram for explaining the load addition in the time direction. In FIG. 5, 5001 to 5003 are 24 Hz video frames, 5004 to 5009 are 60 Hz video frames, and the numerical values attached to the arrows indicate the load values in the time direction when generating the interpolation frames. The interpolation frames 5005 to 5008 are calculated by the above equations (1) and (2), respectively.

  In the memory interface unit 2011, data of the interpolation frame output from the interpolation frame generation unit 2008 is written into the image memory 1004 and data is read at a rate corresponding to the FRC conversion mode 1007, and a frame rate conversion output 2009 is output. Here, as shown in FIG. 5, in the case of conversion from 24 Hz to 60 Hz, the key frame 5002 is deleted and video is output.

  In the interpolation frame generation unit 2008, in addition to the interpolation frame generation operation in the normal FRC operation (hereinafter referred to as vector FRC) as described above, a frame as shown in FIG. Interpolation frame generation by repetition, that is, switching of vector FRC non-operation is performed. FIG. 7 shows an example of the configuration of the interpolation frame generation unit 2008 in FIG. In FIG. 7, 7001 is the current frame signal, 7002 is the previous frame signal, 7003 is the motion vector, 7004 is the FRC conversion mode signal, 7005 is the horizontal / vertical direction interpolation pixel selection unit, 7006 is the time direction interpolation processing unit, and 7007 is A selector section 7008 is a determination signal.

  Here, we will explain the opinions we have obtained through the video quality evaluation by frame rate conversion.

  In the case of frame rate conversion of integer multiples, such as conversion from 60 Hz to 120 Hz, 30 Hz to 60 Hz, and conversion in which only one interpolation frame is inserted between key frames, the image quality of the inserted interpolation frame is a motion vector error. Even if some deterioration is observed in detection or the like, it is difficult for the human eye to recognize it, and as a result, the impact of improving the time resolution is greater. On the other hand, when a plurality of interpolated frames are inserted between key frames by non-integer multiple frame rate conversion, such as conversion from 24 Hz to 60 Hz, or 50 Hz to 60 Hz, the image quality of the interpolated frame becomes a motion vector error. When deterioration is detected by detection or the like, two or more deteriorated images continue in terms of time, which is recognized by human eyes, and deterioration of image quality becomes more conspicuous than the effect of improving time resolution.

  Next, a description will be given of a degraded image caused by the erroneous motion vector detection.

  In motion vector detection, as described above, the correlation is basically calculated from the current frame image and the image one frame before by the difference value matching calculation in block units and pixel units, and the position with the highest correlation is calculated. The pixel value of the interpolation frame is calculated from the pixel value. However, as shown in FIG. 8, for example, in the case where the target moving object passes through the back of the obstacle between frames, a part of the target object does not exist in the video for several frames. It cannot be calculated. Usually, in motion vector detection, in order to improve the reliability of the vector, a method of referring to a motion vector detected in the surroundings or a motion vector of the entire screen is often used. This makes it possible to obtain motion vectors with a high degree of accuracy for images in which the entire screen is panned in a certain direction, but accurate motion vector detection is possible for images with multiple different motions on the screen. Suddenly becomes difficult. Furthermore, even in the case of a fast motion that exceeds the motion vector search range, accurate vector detection cannot be performed, leading to image failure. The simplest measure is to expand the motion vector search range, but the possibility of erroneous detection increases accordingly, directly increasing the amount of computation and increasing the circuit scale when considering hardware implementation. To do.

  In the present embodiment, this problem is solved by switching the interpolation method according to the feature of the video. That is, the video interpolation method is dynamically changed based on the motion vector histogram information for each frame. Hereinafter, in the present invention, this method is referred to as dynamic FRC.

  FIG. 9 shows a state diagram of dynamic FRC processing in the present invention. In FIG. 9, S101 is a vector FRC operation state in which an interpolation frame is generated using a motion vector, and S102 is a non-FRC operation state in which an interpolation frame is generated by frame repetition.

  The state transition (S103, S104) is performed by the motion feature determination in the image feature detection unit 2006.

  The image feature detection unit 2006 detects the feature of the motion based on the histogram information based on the detected motion vector, and outputs a determination signal 2007. FIG. 10 shows an example of the vector histogram detection result. FIG. 10 shows a case where the vector detection range in the motion vector detection unit 2004 is set to 5 vertical blocks and 11 horizontal blocks. FIG. 11 shows a graph of the histogram result of FIG. As can be seen from FIG. 11, it can be seen that many motion vectors are concentrated around the vector (0,0) in this example. The image feature detection unit 2006 determines the feature of motion using such a vector histogram distribution.

  For example, when the motion vector histogram distribution is concentrated in an arbitrary direction as shown in FIG. 12, it is determined that there is a high possibility that the entire screen is scrolled due to the dominant motion in the screen. Also, as shown in FIG. 13, when the screen is concentrated on the boundary portion of the search range by a predetermined threshold or more, it is determined that there is a high possibility that there are many fast-moving objects exceeding the search range. Another judgment method is that when the motion exceeds the search range, there is no block that can be matched accurately, and it is expected that there are multiple blocks with the same matching value in the search range. To do. For example, when the matching values are the same, when an algorithm that prioritizes the vector (0,0) or a vector close thereto is used, the distribution is concentrated on the vector (0,0) as shown in FIG. It is possible to do. Therefore, in this case, when the distribution is concentrated on the vector (0, 0) above the predetermined threshold, it is determined that there is a motion exceeding the search range. Whether or not to concentrate on the vector (0,0) depends on the algorithm in vector detection, and is not limited to this. Further, as shown in FIG. 15, when a distribution of a predetermined threshold value or more is scattered, it is determined that the image has a plurality of movements.

  The image feature detection unit 2006 outputs a determination signal 2007 indicating that when an image having a vector histogram distribution as shown in FIGS. The determination signal 2007 may be a 1-bit signal of 1/0, for example. In the case of the vector histogram distribution as shown in FIG. 12, it is determined that the reliability of the detected motion vector is high, and 0 is output. In the case of the vector histogram as shown in FIGS. 13 to 15, 1 is output because the reliability of the motion vector is low and the possibility of erroneous detection is high. The selector unit 7007 switches between the interpolated frame based on the vector FRC and the interpolated frame output based on frame repetition according to the determination signal. Therefore, when the determination signal 2007 is 0 (S103), the process proceeds to the vector FRC operation (S101), and when the determination signal is 1 (S104), the process shifts to the non-FRC operation (S102). Also, during the state transition (S103, S104), transition is made when the state in which the determination signal 2007 is 0 (1) continues for a predetermined frame period in order to prevent adverse effects caused by frequent state switching. Alternatively, it may be configured to have a hysteresis characteristic of maintaining the state for a predetermined period after the transition.

  In this embodiment, the feature detection of motion is configured by hardware. However, the present invention is not limited to this. Only the motion vector information is output from the hardware, and the dynamic FRC control is realized by software. It is good also as a structure.

  In addition, the interpolation frame output switching need not be performed by the interpolation frame generation unit 2008, and may be configured to be performed by data read control by the memory interface unit 2011.

  As described above, according to this embodiment, it is possible to capture the characteristics of the motion of the input image and dynamically switch the operation state of the FRC according to the characteristics of the motion.

  FIG. 16 shows a state diagram of dynamic FRC processing of the image processing apparatus according to the second embodiment of the present invention. In FIG. 16, the same components as those shown in FIG. 9 are denoted by the same reference numerals, and the description thereof is omitted.

  In this embodiment, the FRC state is set in three stages, thereby enabling more optimal control for the image. Hereinafter, the operation of portions different from those of the first embodiment described above will be described.

  In FIG. 16, S201 is an intermediate operation state between the vector FRC operation and the non-FRC operation. These three operation states are: when the determination signal 2007 output from the image feature detection unit 2006 according to the transition condition described later is 0, the vector FRC operation (S205), when 1, the intermediate operation (S202, S204), Sometimes transition to non-FRC operation (S203).

  FIG. 17 is a diagram for explaining an example of the intermediate operation in the conversion mode from 24 Hz to 60 Hz in the present invention. In FIG. 17, 17001 to 17003 are input key frames, and 17004 to 17007 are interpolation frames. In the vector FRC operation (S101), the interpolation frames are all interpolated using motion vectors, and in the non-FRC operation (S102), the interpolation frames are all repetitive frames, that is, 17004 and 17005 are 17001, 17006. 17007, the frame 17002 is inserted as it is. In the case of the intermediate operation (S201), as shown in FIG. 17, only the 17005 and 17006 frames are replaced with a repetitive frame or a linear interpolation frame in consideration of the center of gravity position in the time direction. Here, the linear interpolation frame in consideration of the position of the center of gravity in the time direction is generated by adding a load in consideration of the center of gravity in the time direction by the horizontal / vertical vector “0” for the frames 17001 and 17002. As for how the video looks by this method, the smoothness of moving images, which is the original effect of frame rate conversion, is naturally reduced when compared to the case where a perfect interpolation frame is generated. Therefore, it is possible to realize a video that suppresses the failure and leaves an effect.

  In order to switch between the three operation states, the image feature detection unit 2006 outputs a determination signal 2007 indicating that effect. For example, 0/1/2 is output as a 2-bit signal. For the 1/0 determination described in the first embodiment, switching to 0/1/2 is realized by providing threshold control. A control example will be described below with reference to the flowchart of FIG.

  First, regarding the vector concentration degree, the concentration distribution value indicating how many points equal to or greater than the vector count threshold TH1 are counted (F101). With respect to the boundary distribution degree, the boundary distribution value of the number of places equal to or greater than the boundary vector count threshold TH4 in the search range boundary portion is counted (F102).

  Next, the concentration distribution value is compared with the distribution threshold values TH2 and TH3 (F103, 104). Further, the boundary distribution value is compared with boundary threshold values TH5 and TH6 (F105, 106). The vector FRC operation (F109) is performed when the concentration distribution value is TH2 or less and the boundary distribution value is TH5 or less. When the concentration distribution value is not more than TH2 and the boundary distribution value is not less than TH5 and not more than TH6, the intermediate operation (F108) is performed. Further, the intermediate operation (F108) is also performed when the concentration distribution value is from TH2 to TH3 and the boundary distribution is from TH5 to TH6. When the concentration distribution value is greater than TH3, non-FRC operation (F107) is performed. Furthermore, when the concentration distribution value is TH2 or less and the boundary distribution value is TH6 or more, or when the concentration distribution value is TH2 or more and TH3 or less and the boundary distribution value is TH5 or more, non-FRC operation ( F107).

  With this control, the image feature detection unit 2006 outputs a determination signal 0 (vector FRC operation) / 1 (intermediate operation) / 2 (non-FRC operation) 2007.

  FIG. 19 is a diagram for explaining an operation image by the dynamic FRC control. As shown in the figure, the three operating states are dynamically switched. Between the transition from vector FRC operation to non-FRC operation (and vice versa), there is basically an intermediate operation. As a result, it is possible to reduce the uncomfortable feeling of the image when the operation is switched.

In this embodiment, the vector concentration distribution and the boundary distribution are used to divide the state by using three threshold values for each value. However, the number of threshold values is arbitrary, and the concentration level is called Instead of the concept, the concept of dispersion or the boundary distribution may be controlled independently in the horizontal / vertical direction. In addition, as in the first embodiment, hysteresis is applied when transitioning to each operating state. It is good also as a structure which gives a characteristic.

  In addition, regarding the interpolated frame in the intermediate operation, in FIG. 17, two of the four interpolated frames are replaced with a repetitive frame or a linear interpolated frame in consideration of the center of gravity in the time direction. Instead, it is possible to replace all four frames with the linear frame. Of course, it is possible to replace only one sheet or replace three sheets. Depending on the difference in the number of replacements, intermediate operations may be classified and state transitions may be subdivided into three or more.

  Further, the positions of two repeated frames or linear frame replacement are not limited to the positions of 17005 and 17006, and 17004 and 17007 may be replaced. Further, for example, as shown in FIG. 20, a method of switching the number and position of replacements at predetermined intervals or in accordance with image characteristics may be used. For example, using the vector concentration distribution information as described above, in the intermediate operation, as shown in FIG. 21, the number of replacement frames is decreased when the concentration is high, and the number of replacement frames is increased when the concentration is low. Control may be performed.

  As described above, according to the present embodiment, by adding the FRC state to the ON / OFF mode and the three stages having the intermediate mode, it is possible to perform more optimal control for the image.

  FIG. 22 is a block diagram showing an example of the configuration of the image processing apparatus according to the third embodiment of the present invention.

  22, the same components as those shown in FIG. 1 are denoted by the same reference numerals, and the description thereof is omitted.

  According to the present embodiment, it is possible to switch to optimal interpolation control according to the genre of a program such as sports, news, and movies.

  Further, it is possible to switch to the optimum interpolation control according to the wide mode.

  Hereinafter, the operation of parts different from the first and second embodiments described above will be described.

  In FIG. 22, 22001 is EIT (Event Information Table) data, 22002 is an EIT data processing unit, 22003 is an EIT determination signal, 22004 is a wide mode signal, 22005 is a dynamic FRC operation mode selection unit, and 22006 is a dynamic FRC operation mode signal. is there.

  In BS / CS / terrestrial digital television broadcasting, in addition to video / audio / data broadcasting, information related to programs (eg, program name, program content, program broadcast date, program broadcast start time, program broadcast duration, broadcast channel, program genre code) Etc.) are also transmitted superimposed on the radio wave. The BS / CS / terrestrial digital television broadcast receiving apparatus uses the program information called EIT sent from the broadcasting station, and obtains / configures the program information to provide the user with an electronic program guide function. In FIG. 22, the EIT data processing unit 22002 performs classification as shown in FIG. 23, for example, according to the 1-byte program genre code used in the content descriptor in the EIT. In other words, the input content is 1 where there may be many scenes with intense movement such as sports, animation / special effects, movies, dramas, variety, and there are many scenes with intense movement such as news and wide shows. For those that will not, 0 is output as the EIT decision signal 22003.

  Wide mode signal 22004 is a signal that indicates whether 4: 3 video is enlarged and displayed at 16: 9, and video with black bars on the top and bottom is enlarged and displayed. The current mode set by the user is displayed. It is a signal to show. Although the name of the mode differs depending on each manufacturer, in this embodiment, two types of smooth (with enlargement) and full (without enlargement) are considered.

  The dynamic FRC mode selection unit 22005 generates a dynamic FRC operation mode signal 22006 from the EIT determination signal 22003 and the wide mode signal 22004. The dynamic FRC operation mode referred to here indicates switching of threshold values for switching, for example, the three operation modes described in the second embodiment. That is, the operation of the image feature detection unit 2006 in FIG. 2 is determined by this mode setting. In the dynamic FRC operation shown in FIG. 19, the vector FRC operation time increases by increasing the threshold value. For example, by having two types of tables, a setting with a low threshold and a setting with a high setting, the FRC operation time increases with a high setting, so it is easy to produce an effect, but there is a possibility that a failure can be seen. A lower setting will limit the effect, but it will be difficult to see the failure.

  Therefore, based on the above, for example, the dynamic FRC mode selection unit 22005 outputs a dynamic FRC operation mode signal 22006 as shown in FIG.

  According to the EIT determination signal 22003, 1 is output if the program genre has many scenes with a lot of motion, 1 when the program genre has few scenes with a lot of motion, and 0 when the wide mode is smooth.

  The image feature detection unit 2006 that has received the dynamic FRC operation mode signal 22006 switches the setting to a lower setting when the value is 1 and a higher setting when the value is 0.

  As described above, in this embodiment, it is possible to switch the interpolation control according to the genre data of the program viewed by the user and the display mode.

  In the present embodiment, the program genre is classified by only eight kinds of major classifications. However, the present invention is not limited to this, and it may be configured to switch to the middle classification below. In the present embodiment, only two types of control switching have been described. However, the present invention is not limited to this, and a further subdivided configuration may be used. In addition, the threshold value has been described as an example of the control switching item, but the present invention is not limited to this. For example, the control mode may be controlled by changing the hysteresis period of each operation state to change the balance of each operation mode. .

It is a block diagram which shows 1st embodiment of this invention. 2 shows an example of the configuration of an FRC unit in the first embodiment of the present invention. It is a figure for demonstrating the production | generation of the interpolation frame by a block matching method. It is a figure for demonstrating the production | generation of the interpolation frame by a block matching method. It is a figure for demonstrating the load addition of the time direction in a FRC interpolation frame production | generation. It is a figure for demonstrating the interpolation frame production | generation by frame repetition. 2 shows an example of a configuration of an interpolation frame generation unit in the first embodiment of the present invention. It is a figure which shows the example of the video pattern used as a motion vector false detection. 2 shows a state diagram of dynamic FRC processing. It is a figure which shows an example of a vector histogram detection result. It is a figure which shows an example of vector histogram distribution. It is a figure which shows an example of vector histogram distribution. It is a figure which shows an example of vector histogram distribution. It is a figure which shows an example of vector histogram distribution. It is a figure which shows an example of vector histogram distribution. 6 shows a state diagram of dynamic FRC processing according to the second embodiment of the present invention. It is a figure for demonstrating an example of the interpolation frame production | generation in intermediate operation | movement. It is a flowchart of a dynamic FRC process. It is a figure for demonstrating the operation image of dynamic FRC control. It is a figure for demonstrating the production | generation of the interpolation frame in a dynamic FRC process. It is a figure for demonstrating the production | generation of the interpolation frame in a dynamic FRC process. It is a block diagram which shows 3rd embodiment of this invention. It is a figure for demonstrating an example of program genre classification. It is a figure for demonstrating an example of dynamic FRC operation mode classification | category.

Explanation of symbols

1001 ... Input signal, 1002 ... Resolution conversion unit, 1003 ... Frame rate conversion (FRC) unit, 1004 ... Image memory, 1005 ... Timing control unit, 1006 ... Display unit, 1007 ... FRC conversion mode signal, 2001 ... Current frame signal, 2002 ... 1 frame previous signal, 2003 ... FRC conversion mode signal, 2004 ... motion vector detection unit, 2005 ... motion vector, 2006 ... image feature detection unit, 2007 ... determination signal, 2008 ... interpolation frame generation unit, 2009 ... Interpolated frame signal, 2010 ... Frame rate conversion output signal, 2011 Memory interface unit, 3001 ... Current frame signal, 3002 ... 1 frame previous signal, 3003 ... Interpolated frame signal, 3004 ... Target block, 3005 ... Motion vector search range, 3006 ... minimum difference value block pair, 3007 ... motion vector, 4001 ... motion vector search range 1,4002 ... motion vector search range 2,4003 ... interpolation frame 1,4004 ... interpolation frame 2, 4005 ... Target block 1, 4006 ... Target block 2, 4007 ... Motion vector 1, 4008 ... Motion vector 2, 5001-5003 ... 24Hz video frame, 5004-5009 ... 60Hz video frame, 6001-6003 ... 24Hz Video frame, 6004 to 6009 ... 60Hz video frame, 7001 ... Current frame signal, 7002 ... Previous frame signal, 7003 ... Motion vector, 7004 ... FRC mode signal, 7005 ... Horizontal / vertical interpolation pixel selection unit, 7006 ... Time Direction interpolation processing unit, 7007 ... F selector unit, 7008 ... judgment signal, S101 ... Vector FRC operation state, S102 ... Vector FRC non-operation state, S103, S104 ... State transition, S201 ... Intermediate operation state, S202 to S205 ... State transition , 17001 to 17003 Input key frame, 17004 to 17007 Interpolated frame, 20001 to 20002 ... Input key frame, 20003 to 20006 ... Interpolated frame, 22001 ... EIT (Event Information Table) data, 22002 ... EIT data processing unit, 22003 EIT judgment signal , 22004 ... Widemo De signal, 22005 ... dynamic FRC operation mode selection unit, 22006 ... dynamic FRC operation mode signal.

Claims (6)

An input video signal including a plurality of frames is input, an interpolation frame is generated based on the frame of the input video signal, and a frame including a part of the plurality of frames included in the input video signal and the interpolation frame A video conversion device for converting into a video signal having a sequence,
A motion vector detection unit for detecting a motion vector between a plurality of frames included in the input video signal;
A video feature determination unit that determines the feature of the video by creating histogram information about the motion vector detected by the motion vector detection unit;
A first interpolation frame generation process for generating an interpolation frame using the motion vector detected by the motion vector detection unit and a plurality of frames included in the input video signal; and a plurality of frames included in the input video signal; A second interpolation frame generation process in which one of the frames is directly used as an interpolation frame, and an interpolation frame is generated by using a linear interpolation frame of two frames among a plurality of frames included in the input video signal as an interpolation frame. the third performs the interpolation frame generation processing, the interpolation frame sequence comprising an interpolation frame part of the frame generated by the first to the third interpolation frame generation processing of a plurality of frames included in the input video signal An interpolation frame sequence generation unit for generating
The interpolation frame sequence generation unit is configured to determine a part of the plurality of frames included in the interpolation frame generated in the first interpolation frame generation process and the input video signal according to the determination result determined by the video feature determination unit. Including a first frame sequence generation process for generating a frame sequence including frames, an interpolation frame generated by the second interpolation frame generation process, and a part of a plurality of frames included in the input video signal A second frame sequence generation process for generating a frame sequence; an interpolation frame generated by the first interpolation frame generation process; a part of a plurality of frames included in the input video signal; and the third interpolation frame image conversion device and switches the third frame sequence generation process for generating a sequence of frames including the generated interpolation frame generation process.   In the interpolation frame sequence generation unit, the motion of the motion vector having the number of motion vectors having a count number equal to or greater than a first threshold and the number of motion vectors having a count number equal to or greater than a second threshold. The switching of the first to third frame sequence generation processing is performed according to a determination result determined according to a number in a boundary portion of a motion detection range of a vector detection unit. Video conversion device. An input video signal including a plurality of frames is input, an interpolation frame is generated based on the frame of the input video signal, and a frame including a part of the plurality of frames included in the input video signal and the interpolation frame A video display device for converting and displaying video signals having columns,
A motion vector detection unit for detecting a motion vector between a plurality of frames included in the input video signal;
A video feature determination unit that determines the feature of the video by creating histogram information about the motion vector detected by the motion vector detection unit;
A first interpolation frame generation process for generating an interpolation frame using the motion vector detected by the motion vector detection unit and a plurality of frames included in the input video signal; and a plurality of frames included in the input video signal; A second interpolation frame generation process in which one of the frames is directly used as an interpolation frame, and an interpolation frame is generated by using a linear interpolation frame of two frames among a plurality of frames included in the input video signal as an interpolation frame. the third performs the interpolation frame generation processing, the interpolation frame sequence comprising an interpolation frame part of the frame generated by the first to the third interpolation frame generation processing of a plurality of frames included in the input video signal An interpolation frame sequence generation unit for generating
A display unit for displaying a video signal having an interpolation frame sequence generated by the interpolation frame sequence generation unit;
The interpolation frame sequence generation unit is configured to determine a part of the plurality of frames included in the interpolation frame generated in the first interpolation frame generation process and the input video signal according to the determination result determined by the video feature determination unit. Including a first frame sequence generation process for generating a frame sequence including frames, an interpolation frame generated by the second interpolation frame generation process, and a part of a plurality of frames included in the input video signal A second frame sequence generation process for generating a frame sequence; an interpolation frame generated by the first interpolation frame generation process; a part of a plurality of frames included in the input video signal; and the third interpolation frame video display device and switches the third frame sequence generation process for generating a sequence of frames including the generated interpolation frame generation process. In the interpolation frame sequence generation unit, the motion of the motion vector having the number of motion vectors having a count number greater than or equal to a first threshold and the number of motion vectors having a count number greater than or equal to a second threshold. according to the determination result of the determination in accordance with the number at the boundary portion of the motion detection range of the vector detection unit, according to claim 3, characterized in that for switching the first to third frame sequence generation process Video display device. An input video signal including a plurality of frames is input, an interpolation frame is generated based on the frame of the input video signal, and a frame including a part of the plurality of frames included in the input video signal and the interpolation frame A video conversion method for converting into a video signal having columns,
A motion vector detection step of detecting a motion vector between a plurality of frames included in the input video signal;
A video feature determination step of creating histogram information about the motion vector detected in the motion vector detection step and determining a video feature;
A first interpolation frame generation process for generating an interpolation frame using the motion vector detected in the motion vector detection step and a plurality of frames included in the input video signal; and a plurality of frames included in the input video signal; A second interpolation frame generation process in which one of the frames is directly used as an interpolation frame, and an interpolation frame is generated by using a linear interpolation frame of two frames among a plurality of frames included in the input video signal as an interpolation frame. the third performs the interpolation frame generation processing, the interpolation frame sequence comprising an interpolation frame part of the frame generated by the first to the third interpolation frame generation processing of a plurality of frames included in the input video signal An interpolated frame sequence generating step for generating
In the interpolation frame sequence generation step, the interpolation frame generated in the first interpolation frame generation process and a part of the plurality of frames included in the input video signal are determined according to the determination result determined in the video feature determination step. Including a first frame sequence generation process for generating a frame sequence including frames, an interpolation frame generated by the second interpolation frame generation process, and a part of a plurality of frames included in the input video signal A second frame sequence generation process for generating a frame sequence; an interpolation frame generated by the first interpolation frame generation process; a part of a plurality of frames included in the input video signal; and the third interpolation frame and switches the third frame sequence generation process for generating a sequence of frames including the generated interpolation frame generation process Image conversion method. In the interpolation frame sequence generation step, the number of motion vectors having a count number equal to or greater than a first threshold value in the entire histogram and a motion vector having a count number equal to or greater than a second threshold value are detected in the motion vector detection step. range according to the image feature judging step of judging a result of the determination as a function of the number of boundary portions, according to claim 5, characterized in that for switching the first to third frame sequence generation process Video conversion method.

Images